Descriptive results
The descriptive analysis was as follow. The features for gender and age of FDS and the general public were presented in Table 1. The average age of FDS was approximately 13 years younger than that of the general public group.
Table 1. The Gender and age distributions among FDS and the general public
Variable
|
FDS
|
General Public
|
In total(n)
|
269
|
258
|
Gender (Percentage)
|
|
|
Male
|
85 (31.60%)
|
100(38.76%)
|
Female
|
184 (68.40%)
|
158 (61.24%)
|
Age
|
|
|
Range
|
20-56
|
23-56
|
Average (SD)
|
27.20 (8.40)
|
40.53 (10.00)
|
As for some potential factors, a rather heavy workload was perceived among FDS. In Table 2, the majority of FDS worked no more than 5 days per week, whereas 28.25% of them worked longer. There were 11.89% of the FDS worked more than 8 hours a day and 29.00% of the FDS continued working for more than 2 hours before a break.
Then, the average score of FDS on the COVID-19 knowledge was 2.46 (SD = 1.33). The percentage of the correct answer for each question was shown in Table 3. FDS were more likely to choose wrong answer on effective sanitary chemicals, the main routes for transmission and the discharging criteria.
Table 2. Descriptive results of workload among FDS.
Variable
|
Frequency (%)
|
Total
|
269(100.00%)
|
Working days per week
|
|
0-4 days
|
7(2.60%)
|
5 days
|
186(69.15%)
|
6-7 days
|
76(28.25%)
|
Working hours per day
|
|
<4 hours
|
35(13.01%)
|
[4, 6) hours
|
49(18.22%)
|
[6, 8) hours
|
153(56.88%)
|
≥ 8 hours
|
32(11.89%)
|
Working hours between breaksa
|
|
<1 hour
|
68(25.28%)
|
[1, 2) hours
|
123(45.72%)
|
[2, 4) hours
|
59(21.93%)
|
≥ 4 hours
|
19(7.07%)
|
a time intervals less than 5 minutes were not counted as a break
Table 3. Descriptive results of knowledge of COVID-19 among FDS.
Questions
|
Frequency (%)
|
Total
|
269(100.00%)
|
the common symptoms
|
|
correct
|
221(82.16%)
|
incorrect
|
48(17.84%)
|
the incubation period
|
|
correct
|
158(58.74%)
|
incorrect
|
111(41.26%)
|
the main routes of transmission
|
|
correct
|
91(33.83%)
|
incorrect
|
178(66.17%)
|
cleaning and disinfectiona
|
|
correct
|
68(25.28%)
|
incorrect
|
201(74.72%)
|
the discharging criteria
|
|
correct
|
41(15.24%)
|
incorrect
|
228(84.76%)
|
the susceptibility of peopleb
|
|
correct
|
83(30.86%)
|
incorrect
|
186(69.14%)
|
athe effective measures and chemicals for cleaning and disinfection
bthe susceptibility of people to COVID-19
Comparison of anxiety state between FDS and the general public
As the result showed, 60 out of 269 FDS (22.30%) suffered from anxiety disorders and 16 out of 258 ordinary people (6.20%) suffered from anxiety disorders. there was a significant association (χ² = 27.671, p <0.01) between being FDS and having anxiety disorders. Based on the odds ratio, the likelihood of FDS to suffer from anxiety disorder was 4.342 (95% CI: 2.427-7.768) times higher than that of the general public.
Bivariate analysis
As for bivariate analysis, firstly, the associations of several potential factors (categorical variables) with the anxiety state of FDS were assessed. As was shown in Table 4, factors that were significantly associated with the outcome variable (the anxiety state of FDS) included social relationships with colleagues and patients, the treatment to suspected or confirmed cases, and adopting PM-3. But the treatment to suspected or confirmed cases only had a weak association with the outcome variable, since the Cramer’s V was 0.139.
Table 4. Associations between potential factors and anxiety disorders of FDS (categorical variables).
Variable
|
Anxiety state
|
χ²
|
Cramer's V
|
P- value
|
|
Yes
|
No
|
|
|
|
|
(n=60)
|
(n=220)
|
|
|
|
Gender
|
|
|
2.441
|
0.095
|
0.118
|
Male
|
14
|
71
|
|
|
|
Female
|
46
|
138
|
|
|
|
Aerosolization proceduresa
|
|
|
0.009
|
0.006
|
0.926
|
Yes
|
44
|
152
|
|
|
|
No
|
16
|
57
|
|
|
|
Conflicts with colleagues and/or patientsb
|
|
|
13.833
|
0.227
|
<0.001
|
Yes
|
14
|
14
|
|
|
|
No
|
46
|
195
|
|
|
|
Treatment to suspected or confirmed casesc
|
|
|
5.200
|
0.139
|
0.023
|
Yes
|
11
|
17
|
|
|
|
No
|
49
|
192
|
|
|
|
Exposure to potential infectious substance d
|
|
|
3.252
|
0.110
|
0.071
|
Yes
|
18
|
40
|
|
|
|
No
|
42
|
169
|
|
|
|
PM-3e
|
|
|
10.246
|
0.195
|
0.001
|
Yes
|
43
|
185
|
|
|
|
No
|
17
|
24
|
|
|
|
a Did you often perform aerosolization procedures?
b Did you have conflicts with colleagues and/or patients in the last six months?
c Have you performed treatment to suspected or confirmed cases of COVID-19?
d whether your skin or wounds were exposed to the blood, saliva, or other body fluids of patients?
e Does your workplace follow PM-3 guideline?
In addition, other factors (numerical variables) were also listed as follow. Table 5 showed the Eta squared of each predictor variable. It was shown that there were four factors (a. the number of working days per week, b. the number of working hours per day, c. their knowledge of COVID-19, d. the number of trainings organized by workplace) that had no or weak association with the anxiety state of FDS. But the age factor had a moderate association.
Table 5. Associations between potential factors and anxiety disorders of FDS (numerical variables).
Variable
|
η²
|
Age
|
0.217
|
The number of working days per week
|
0.042
|
The number of everyday working hours per day
|
0.028
|
The number of hours between breaks
|
0.014
|
Knowledge of COVID-19
|
0.128
|
Binary logistic regression
A binary logistic regression model was then developed to see the relations between the predictor variables and outcome variable (the anxiety state of FDS). The Chi-square for this model was 46.473 (p < 0.001), suggesting that the model fitted the data well. The overall classification accuracy rate of the model was 81.78%. Specifically, 95.69% of the FDS who have no anxiety disorders were accurately classified, but the accuracy rate was lower for FDS who had anxiety disorders (33.33%). The coefficients of all predictor variables were presented in Table 6. These three predictor variables were significantly related to the anxiety state of FDS and predictive to the outcome variable. With the increase of age, FDS were less likely to suffer from anxiety disorders. Similarly, applying PM-3 at dental settings would predict the reduction of FDS with anxiety disorders. On the contrary, FDS who had conflicts with colleagues and/or patients were more likely to develop anxiety disorders. More specifically, based on the odds ratio, FDS with an elder age were less susceptible to anxiety disorders by a factor of 0.857. The odds of suffering from anxiety disorders was 0.243 times lower for FDS with PM-3 than that of FDS without PM-3. Lastly, FDS who had conflicts with colleagues and/or patients were 2.991 times more likely to suffer from anxiety disorder than who did not.
Table 6. Results of binary logistic regression analysis.
Variable
|
Coefficient
|
OR (95%IC)
|
P-value
|
Age
|
-0.154
|
0.857(0.792-0.925)
|
<0.001
|
PM-3a
|
-1.415
|
0.243(0.105-0.561)
|
0.001
|
Conflicts with colleagues and/or patientsb
|
1.095
|
2.991(1.238-7.227)
|
0.015
|
Constant
|
3.609
|
36.7929
|
0.001
|
a Does your workplace follow PM-3?
b Did you have conflicts with colleagues and/or patients in the last six months?